APPLICANT'S ABSTRACT: The objective of this proposal is to study the impact of paternal alcoholism on outcomes and functioning of male and female adolescent offspring. This aim will be accomplished by comparing data on adolescent offspring of 878 twins who are either concordant or discordant for alcoholism. Twin family studies afford important design advantages: identical twins share all their genes and fraternal twins share half so that genetic contributions to the onset of alcoholism are enrolled. The first aim will determine whether and to what extent the offspring of alcoholic twins compared to the offspring of non-alcoholic twins differ in regard to; alcohol use and abuse, psychopathology, academic achievement, general psyhcosocial competence and family relationships assessed by structured interviews and questionnaires. Second, the study will be extended to a twin-family design by collecting information on biological mothers' alcohol use and relevant maternal and paternal psychopathology. This will clarify differences between the comparative functioning and outcome data for the offspring of alcoholic fathers and offspring of non-alcoholic fathers by examining the role of spousal/maternal effects and twin father psychopathology co-morbid with alcoholism. Maternal effects can be controlled, and analysees will be conducted to assess the role of assortative mating, the tendancy of alcoholic parents to marry each other. Third, in order to understand the mechanisms whereby paternal alcoholism may impact offspring, three mediation models for familial transmission of alcoholism will be tested: deviant socialization, negative affect regulation and pharmacological vulnerability. This proposed research makes use of a unique resource because the twin fathers are all members of the Vietnam Era Twin Registry and have all participated in previous waves of data collection. While the offspring of these fathers are still adolescents, invaluable diagnostic, demogrphic and behavioral data have allowed us to characterize independent variables such as fathers' drinking history, psychopathology, and demographic variables such as income and education and family composition. Therefore, the proposed research gains the advantage of being a second wave investigation in which maximally informative families have already been ascertained.